BACKGROUND German social legislation gives patients the right to obtain a second opinion before elective surgery and defines quality criteria for reimbursement by statutory health insurances. However, the effects of second opinions before elective surgery are largely unknown. The aim of this study was to evaluate the effects of a second opinion programme in patients recommended for knee arthroplasty. METHODS The largest statutory health insurance funds in Bavaria offered patients who had been recommended to have knee arthroplasty the opportunity to partake in a second opinion programme which consisted of an in person presentation to an experienced knee surgeon. In this cohort study, consecutive patients from this second opinion programme who signed informed consent were included from 07/10/2016 to 14/02/2020. Data were collected before and after the second opinion visit. RESULTS A total of 141 (66%) of 215 patients who presented for a second opinion participated in the evaluation study. The second opinion physician recommended knee arthroplasty to 40% of the patients, later knee arthroplasty if the conditions worsened to 40%, and no knee arthroplasty to 20%. After receiving the second opinion 28 of 56 (41%) undecided patients preferred knee arthroplasty, 14 no knee arthroplasty, 14 remained undecided. Four of 46 patients with a preference for \textquotedblarthroplasty\textquotedbl changed their decision to \textquotedblno arthroplasty\textquotedbl, five of 35 patients from \textquotedblno arthroplasty\textquotedbl to \textquotedblarthroplasty\textquotedbl. The patients were more confident in their decision according to the decision confidence scale (before: 5.4 ± 3.0; after: 7.8 ± 2.5; p~< 0.001). They rated their satisfaction with the second opinion programme with a mean grade of 1.35 (± 0.60) (best:1; worst:6). Logistic regression analyses showed that the recommendation of the second opinion physician for joint arthroplasty was associated with the guideline criteria radiological severity of osteoarthritis (p~= 0.001) and knee-joint-specific quality of life (p~= 0.041). CONCLUSION The second opinion of an experienced knee surgeon frequently deviates from the initial recommendation for knee arthroplasty. The association of guideline criteria to the second recommendation suggests a high quality of the second opinion. From the patient perspective, the second opinion reduces uncertainties in their treatment decision.